Teradata
The target system user must have read and write access to the Teradata target system. The following example shows how to grant read and write privileges on the target system.
Grant select on DBC.ColumnsV to {User}; Grant select on DBC.AllRightsV to {User}; Grant select on DBC.UDTInfo to {User}; Grant select on DBC.TablesV to {User};
For the data field, Teradata supports CLOB, JSON, or VARCHAR. If you use VARCHAR and send a message larger than the preallocated column size, the message is truncated.
HDFS
The target system user must have permission to create and modify files in HDFS for the defined data path. HDFS includes the following default configurations that are not adjustable:
Configuration | Value |
---|---|
File Rotation Size | 20 MB |
File Rotation Time | 1 Day |
Block Size | 128 MB |
Buffer Size | 4 kb |
Replication Factor | 3 |
File Naming Conventions | /user/defined/path/[target_id]/file.seq |
Compression Code | Sequence |
Before Listener can write data to an HDFS target for a Kerberized Hadoop instance, valid credentials for the following are required:
- KDC
- Principal
- Base64 encoded keytab file content
The following example shows the fields where the credentials must be provided.
...
"data_path": {
"extension": "seq",
"path": "/user/testuser/kerberos"
},
"properties":{
"kdc":"ip-xxx-xx-xx-xxx.location.compute.internal:88",
"principal":"testuser@ABC.LOCATION.COMPUTE.INTERNAL",
"base64KeytabContent":"BQIAAABdAAE...F+zcAAAAB"
}
}' \
-i \
https://CLUSTER.DOMAIN/listener/appservices/targets
HBase
The target system user must have permission to access and write to the table and column family.